Abstract
In practice, time series forecasting involves the creation of models that generalize data from past values and produce future predictions. Moreover, regarding financial time series forecasting, it can be assumed that the procedure involves phenomena partly shaped by the social environment. Thus, the present work is concerned with the study of the use of sentiment analysis methods in data extracted from social networks and their utilization in multivariate prediction architectures that involve financial data. Through an extensive experimental process, 22 different input setups using such extracted information were tested, over a total of 16 different datasets, under the schemes of 27 different algorithms. The comparisons were structured under two case studies. The first concerns possible improvements in the performance of the forecasts in light of the use of sentiment analysis systems in time series forecasting. The second, having as a framework all the possible versions of the above configuration, concerns the selection of the methods that perform best. The results, as presented by various illustrations, indicate, on the one hand, the conditional improvement of predictability after the use of specific sentiment setups in long-term forecasts and, on the other, a universal predominance of long short-term memory architectures.
Highlights
The observation of the evolution of various time-dependent phenomena, as well as the decision-making based on structures predicting their future behavior have greatly shaped the course of human history
The first section deals with the utilization of textual data in light of sentiment analysis for the task of time series forecasting and the investigation of whether or not and when their use has a beneficial effect on improving predictions
The overall structure and results of an extensive experimental procedure were presented, in which 22 different input configurations were tested, utilizing information extracted from social networks, in a total of 16 different datasets, using 27 different algorithms
Summary
The observation of the evolution of various time-dependent phenomena, as well as the decision-making based on structures predicting their future behavior have greatly shaped the course of human history. Time series forecasting is the process in which the future values of a variable describing features of a phenomenon are predicted based on existing historical data using a specific fit abstraction, i.e., a model All such time-dependent features containing past observations are represented as time series. Depending on the number of attributes, i.e., the dependent variables describing observations recorded sequentially over the predefined time steps, whose values are collected at any given time, a distinction is made between univariate and multivariate time series [1]. Such methods find application in a wide range of time-evolving problems. The domain has flourished in recent decades, as the demand for better and better models remains increasingly urgent, as their use can greatly contribute to the optimization of decision-making and lead to better results in various areas of human interest
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
